Our long term goal is to understand the neuronal substrate of the pulse generator and cyclic trigger that regulate gonadotropin secretion. Our current hypothesis is that plasticity of synaptic organization onto the LHRH neuron might mediate changes in the pulsatile and cyclic mode of release. Both negative and positive feedback effects of steroids as well as information about the external environment might be conveyed not only by changes in synaptic activity but in synaptic arrangement as well. We propose to (1) determine which LHRH neurons project to the median eminence and which to extrahypothalamic areas. Once we know the population that forms the final common pathway for gonadotropin release we shall (2) quantify the type and distribution of synapses onto these cells in animals that differ in their sensitivity to the negative feedback effects of gonadal steroids. Our preliminary data suggest that at least some of the LHRH neurons that project to the median eminence are local circuit neurons. We propose to study this question. This is particularly important since the same cell that has a neurosecretory terminus may well be influencing its neighbors via more classical synaptic output. Both of these latter experiments will utilize ultrastructural immuocytochemistry. since there is substantial information on the roles of monoamines and beta-endorphin on either pulsatile or cyclic gonadotropin release, we shall determine the distribution of synaptic terminals containing these neurotransmitters relative to LHRH cells, dendrites and particularly neurosecretory terminals. This study will combine immunocytochemistry with radioautography at the ultrastructural level. Finally we will look at the intrahypothalamic input to LHRH neurons and determine if cells of the arcuate nucleus and/or the suprachiasmatic nucleus synapse on LHRH neurons. This study will combine anterograde degeneration following a lesion with immunocytochemistry. These ultrastructural studies are designed to give a better understanding of the way in which the CNS modulates reproductive function. We shall use rats, guinea pigs or hamsters. In some experiments we will employ a comparative approach but in most experiments a particular species is chosen because it represents a good model for a specific endocrine condition.